G Model
FSI-6081; No. of Pages 6 Forensic Science International xxx (2010) xxx–xxx
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
The investigation of a relative contrast index model for fingerprint quantification Jana Vanderwee a, Glenn Porter a,*, Adrian Renshaw a, Michael Bell b a b
School of Natural Sciences, University of Western Sydney, Locked Bag 1797, Penrith South DC 1797, Australia NSW Police Forensic Services Group, Australia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 13 July 2009 Received in revised form 3 May 2010 Accepted 11 May 2010 Available online xxx
The quantification of fingerprint contrast is a relatively new concept in fingerprint enhancement research. It has emerged as a mode of fingerprint assessment to reduce the potential biased of visual qualitative assessment. Subjective qualitative methods that are currently reported in the literature include; side-by-side assessment, assigning a score to a treatment based on visible criteria and stating observed results without presenting supporting validation. These qualitative methods often do not state clearly the visual assessment parameters and produce a degree of ambiguity when defining the enhancement results. The relative contrast index model was constructed to empirically quantify the difference in contrast between fingerprint ridges and valleys, using measurements gained from a microspectrophotometer. This paper seeks to further investigate this recent research and test the model using three different microspectrophotometers. Data from these separate sources will determine whether the theoretical aspects of the model would pragmatically produce reliable and repeatable results across a range of microspectrophotometers found in forensic laboratories. ß 2010 Elsevier Ireland Ltd. All rights reserved.
Keywords: Fingermark enhancement Contrast Fingerprinting Photography Optical enhancement Relative contrast index
1. Introduction Latent fingermark enhancement is an active research arena. Enhancement techniques that are routinely used may not always be ideal for all surfaces, may be ineffective for weak latent traces and may not reveal enough detail for identification. Particular areas of ongoing fingermark research include; the optimisation of new fingermark detection techniques, new powder formulations using dyes and emerging nanotechnology, employing new optical techniques and digital image processing [1]. A survey of the fingerprinting literature revealed that several qualitative methods are currently used in research to describe fingermark development and enhancement results [1–46]. A review of the literature was conducted using specific criteria to determine the type, extent and frequency of fingermark assessment methods and categorised them according to the type of enhancement researched (reagent-based, metal deposition, powders, latent blood and other). The various methods of describing fingermark enhancement results are shown in Fig. 1. The key criteria during the literature review were the number of fingermark enhancement images that were presented and
* Corresponding author. Tel.: +61 2 4570 1739. E-mail address:
[email protected] (G. Porter).
classifying any visual assessment as ‘defined’,1 ‘not clear’ 2 or ‘undefined’ 3 depending on how clearly the assessment parameters were stipulated in each paper. Qualitative visual assessments of results were made across all articles within the review. Furthermore, it was noted if any quantification of the resultant fingermark enhancement was attempted and if so, what type of quantification was used. The frequency of visual comparisons, visual scores, side-by-side comparisons and the numbers developed were also determined. The production cost and development time were also factors that were not specifically related to the enhancement but do numerically quantify an aspect of the method. Table 1 illustrates the results of the literature review. The quantification of contrast has recently emerged as a numerical alternative to the subjective and often ambiguous
1 ‘Defined’ category was used when the article clearly explained the parameters used for enhancement evaluations. For example ‘‘Stained prints were analysed for level I and level II detail. If the ridge flow or pattern of the fingerprint was identifiable, it was considered to posses level I detail. The presence of level II detail included the observation of bifurcations, ridge endings, a clear core area or one or more deltas’’ [28]. 2 ‘Not clear’ category is defined when parts of the qualitative assessment are explained, however, not made completely clear. 3 ‘Undefined’ category was used when results were described without any explanation regarding the assessment parameters. For example; ‘‘Staining of fingerprints was patchy and less dark’’ [30] and ‘‘Excellent definition was seen in both cases’’ [26].
0379-0738/$ – see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2010.05.005
Please cite this article in press as: J. Vanderwee, et al., The investigation of a relative contrast index model for fingerprint quantification, Forensic Sci. Int. (2010), doi:10.1016/j.forsciint.2010.05.005
G Model
FSI-6081; No. of Pages 6 J. Vanderwee et al. / Forensic Science International xxx (2010) xxx–xxx
2
2.1. Instrument 1 – Leica DMR and Ocean Optics HR2000 Instrument 1 was a non-dedicated instrument that linked an Ocean Optics HR2000 spectrophotometer by a coaxial probe to the camera mount of the Leica DMR microscope. The OOIBase32 software program version 2.0.6.5 was opened to process the spectra. The spectrophotometer had fixed specifications with the entrance aperture set at 50 mm, slit width of 25 mm, optical resolution of 0.1 nm, composite diffraction grating that ranged 200 nm to 1100 nm. The integration time was set to 10 ms. The boxcar and averaging functions were each set to 5. All other functions were either set to 0 (zero) or were not selected. The microscope lamp brightness was set to the maximum and the lamp was allowed to warm up for a minimum of 60 min before use. Any light filters on the microscope were all disengaged. The total magnification was 400 with the combination of the eyepiece (10) and the objective lens (40). A large magnification was used to compensate for the large sampling area read by the spectrophotometer.
Fig. 1. Methods of describing fingermark enhancement results used in current literature.
qualitative assessment methods. The relative contrast index (RCI) measures the difference in contrast between fingerprint ridges and valleys using a microspectrophotometer [47]. It provides a numerical or logarithmic value of contrast that is comparable between treatments. It also eliminates visual effects from influencing the visual assessment of a specific technique. Relative contrast index ¼ log10
Valley intensity Ridge intensity
The relative contrast index method uses spectra from fingermark valleys and ridges using a microspectrophotometer and the numerical values required for the relative contrast index are obtained by integrating the area under the spectral curve and applying the model’s formulae. In theory, due to the relative nature of the contrast index, the model should produce consistent results for the same specimen when using different instruments. The principal objective of this research was to examine the relative contrast index and determine whether these values are consistent using different microspectrophotometers found within forensic science laboratories. 1.1. Measurement mode The initial microspectrophotometer used in the development of the relative contrast index model [47] had the capacity to measure spectra in ‘reflective’, ‘transmission’ and ‘scope’ modes. The initial RCI model was developed using ‘scope mode’ which is a measurement mode that uses ‘intensity counts’ as the measurement of reflected light detected. The dedicated instruments used in this research did not have the ‘scope’ mode capacity. ‘Percentage reflection’ mode was considered as a more viable measurement mode. A preliminary experiment for this research required testing the original instrument in both ‘scope’ and ‘percentage reflection’ modes to determine whether the relative contrast index results would be equivalent. Spectra were measured in both ‘scope’ and ‘percentage reflection’ modes, keeping all other settings identical including the sample area. The difference between the scope and percentage reflection modes was considered negligible but for the purpose of this study, all microspectrophotometer instruments were utilised in ‘percentage reflection’ mode to ensure a standardised approach. 2. Materials and methods Three different microspectrophotometers were selected to measure the experimental sample material and determine whether the relative contrast index would remain the same throughout each instrument. The following instruments and operational parameters were used:
2.2. Instrument 2 – Leica Aristomat and Leitz MVP SP The Leica Aristomat microscope was fitted with a dedicated Leitz MVP SP spectrophotometer. This instrument was also formerly used by law enforcement for forensic casework. The Leica Spectra Program, version 1.32 for Windows 95, was opened to process the spectra. In the adjustment window, the sensitivity was set at 7.5%, microscope lamp brightness was set at 50% and the ‘active’ box had the eyepiece flaps box checked. The lamp shutter key was selected for all the measurements. The low pass box had the 3000 Hz low pass filter selected. All optical filters were disengaged by selecting ‘open’. The ‘equalize’ box was left blank. In the measurement window, the miscellaneous box had the spectral option ticked. The interval was set at the lowest value possible, 0.1 s. Display also had spectrum selected. Colorimetry values had light type A and colour model XYZ selected. The spectrum box had a range of 400 nm to 700 nm, with the limit set to 0 (zero) and delta 3 selected. Measurement mode had ‘reflectance’ selected with smooth set to 5 and number of scans set to 5 (by default 8). Additive was not selected. The photometer field diaphragm had the two levers parallel to the red dot. The microscope lamp was allowed to warm up for a minimum of 60 min before readings were taken. 2.3. Instrument 3 – CRAIC QDI2010 The CRAIC QDI2010 is currently used by law enforcement for forensic casework. CRAIC MSP Data Acquisition Software was opened to process the spectra and CRAIC CCD Image Capture (IC) software (DFx41AF02) was used to view the samples. The optimum integration time was calculated by the instrument at 1913.23 ms. Standard analysis conditions were set with 400 nm to 700 nm selected. Scans to average were set at 20 with the recommended sampling time of 1242.51 ms. The resolution factor (0–15) was set at 4. The video formats had a frame rate of 1280 960 at 7.5, 3.75 frames/s. The dynamic range of the ADC was 10 bit and the signal to noise ratio was ADC 9 bit at 25 8C gain 0. The IC Capture 2.0 was set at 50% live for visualisation. The exposure was set between 1/83 s and 1/120 s for the duration of the data collection. Brightness was set to 63, gain to 300 and auto reference parameter to 690. The colour settings were kept at the optimum values with hue 181, saturation 129 and white balance auto was selected. The image parameter was set at gamma 12. 2.4. Fingermark exemplar material The experimental fingermark samples consisted of an inked depletion series which was used in all microspectrophotometric data acquisition. Inked fingermarks were used due to the time lapse between data collection from each instrument and ink deposition was considered a more stable material than other fingermark development methods (e.g. ninhydrin, amido black, physical developer). The depletion series provided samples of differing contrast across a range of depletions between each sample group. The inked fingermarks were deposited onto Fuji Xerox Performer+1 copy paper using a finger loaded with a Lightning Powder Company Incorporation1 black Porelon Fingerprint Pad. The male donor freshly loaded the finger in ink and then rolled the finger onto the copy paper, ensuring a fully deposited mark was deposited using a rolled nail-to-nail technique. The thumb was freshly inked and then consecutively deposited three times without re-inking to produce three depletions. This was repeated 30 times with a total of 90 different fingermarks (30 each group). The sample fingermarks were labelled according to deposition with the first known as the n1 depletion, followed by the n2 depletion and then the n3 depletion (Fig. 2). The inked samples were then stored in Camerons Premium1 blank envelopes to avoid any fading caused by ultraviolet radiation and further stored in an insulated Valuca Pty Ltd. Arctic 4L Styrofoam cooler to maintain consistent temperature and enhance the archival considerations. 2.5. Reference standard and control A mini GretagMacbeth ColorChecker1 colour rendition test chart was used as a reference standard and control which consists of twenty-four colour patches of
Please cite this article in press as: J. Vanderwee, et al., The investigation of a relative contrast index model for fingerprint quantification, Forensic Sci. Int. (2010), doi:10.1016/j.forsciint.2010.05.005
G Model
FSI-6081; No. of Pages 6 J. Vanderwee et al. / Forensic Science International xxx (2010) xxx–xxx
3
Table 1 Results from literature survey regarding methods of articulating fingermark enhancement results. Enhancement research Reagent
MD
Powder
Blood
Other
Fingermark image\s
Visual assessment
Visual comparison
Visual score assigned
Side-by-side comparison
Development time
*
*
* *
[2] [3] [4] [5] [6] [7]
0 5 2 3 5 6
U U U U U U
* * * * * *
[8]
0
U
*
[9]
0
U
[10] [11] [12] [13] [14] [15]
12 13 0 6 12 25
[16] [17]
9 5
U U
*
[18] [19]
22 7
U U
*
[20] [21]
15 12
U U
* *
* *
[22] [23]
11 3
U U
* *
*
[24] [25]
16 8
U NC
* *
[26]
17
U
*
[27]
8
D
*
[28] [29] [30] [31]
5 6 0 22
D U U U
* * * *
[32] [33]
2 2
U U
[34] [35] [36] [37] [38] [39] [40] [41] [42]
6 0 16 8 5 4 10 34 3
U U U U NC NC U U U
[43] [44] [45]
2 18 21
D U D
[46]
16
NC
U U U U U NC
* *
Production cost
Number of marks developed
Type of quantification
* * *
Spectra Spectra Spectra, amino acid test Spectra, NMR spectra Spectra, fluorescence intensity Spectrum Amino acid test
*
*
* *
* * * * *
Quantification attempted
* * * * *
* *
* *
* * *
*
* *
*
*
* *
Deposition (ICP-MS)
*
Deposition (Densitometry & ICP-MS)
* *
Spectra Weight and volume percent Spectra Spectra, fluorescence intensity, lifetime
*
*
* *
*
*
* * * *
*
*
*
*
*
*
*
Minutiae counted % Success rate
*
Spectra
* *
Cyanoacrylate deposition Spectra
*
Spectra
*
Sputter time, spectra, surface analysis
*
pH testing, Ca spectra, % success rate
* *
*
* *
*
* *
* *
*
Visual assessment key, D = Defined, U = Undefined, NC = Not clear. known colour and reflective values. This study used only the six monochrome references considered as a grey scale of known reflective values. The grey scale consisted of black, white and four shades of grey (Fig. 3). Forty random measurements were taken from each reference standard patch. Data was then analysed to ensure the microspectrophotometers were producing comparable results for the reference standards.
2.6. Data collection and analysis All data collection parameters were maintained upon the specific instrument for the entire sample collection period to ensure consistency. Each fingermark sample had ten valley readings and ten ridge readings taken on each instrument per group. These spectra were then imported into Microsoft1 Excel and this
Please cite this article in press as: J. Vanderwee, et al., The investigation of a relative contrast index model for fingerprint quantification, Forensic Sci. Int. (2010), doi:10.1016/j.forsciint.2010.05.005
G Model
FSI-6081; No. of Pages 6 J. Vanderwee et al. / Forensic Science International xxx (2010) xxx–xxx
4
Fig. 2. Inked fingermark depletion sample.
Fig. 3. Gretag Macbeth Color Checker1 reference standard.
software was used to construct the relative contrast indices and conduct statistical analysis. The relative contrast index model analysed the spectral data produced in ‘scope’ mode. The data analysis used the spectral data determined by integrating the area under the curve [47]. It was considered that averaging the ‘percentage reflection’ spectral data might be a more meaningful way of analysing the linear response. However, it should be noted that coloured values do not produce a similar curve response. The data obtained in the visible region was integrated (with values below 3% omitted as white noise) and then compared to the same data mean values (averaging). The raw numerical values were considerably different (Table 2). However, when these values were entered into the relative contrast index formula the results were identical. These results demonstrate that either data analysis method can be used to enter into the relative contrast index formula. A one way ANOVA was completed on the depletion series samples to determine whether the relative contrast index could be used to distinguish between each depletion group (n1, n2, n3).
3. Results
the third (Fig. 5). All three instruments exhibited this trend. There was no overlap observed between the deviations of the different depletions on any of the instruments (Table 3). The data spread reduces similarly from n1 depletion to n3 depletion, with the data spread most in the n1 depletion, less in the n2 depletion and least in the n3 depletion which represents a lowering of contrast. One way ANOVA analysis of the results indicated there was a significant difference between the mean RCI values of Instruments 1, 2 and 3 for each depletion series (Instrument 1, F2,87 = 367.4, P < 0.0001; Instrument 2, F2,87 = 262.2, P < 0.0001 and Instrument 3, F2,87 = 437.3, P < 0.0001). There was also a significant difference between the means of the n1 depletion, n2 depletion and n3 depletion on each instrument (n1 depletion, F2,87 = 86.5, P < 0.0001; n2 depletion, F2,87 = 115.3, P < 0.0001 and depletion 3, F2,87 = 119.7, P < 0.0001).
3.1. Reference standard
4. Discussion
Each grey value from the reference standard was measured and discriminated according to the different tonal values. However, equivalent percentage reflection values were not achieved from the three experimental instruments (Fig. 4). Instrument 2 almost halved the relative contrast index values obtained per grey tone. Contrastingly, Instrument 1 and 3 decreased more gradually.
The relative contrast index model was designed to be simple to implement. Irreconcilable differences existed between the instruments that even the relativity of the model could not compensate
3.2. Fingermark exemplar samples The n1 depletion produced the highest relative contrast index, n2 depletion produced the second highest values and n3 depletion Table 2 Relative contrast index results from each fingermark depletion group. Depletion
Data analysis
Average ridge
Average valley
RCI
n1
Integration Averaging
19035.73 28.63
52862.13 79.49
0.4436 0.4436
n2
Integration Averaging
30520.95 45.9
53011.68 79.72
0.2398 0.2398
n3
Integration Averaging
38098.11 57.29
54723.11 82.29
0.1573 0.1573
Fig. 4. Relative contrast index values obtained from the reference standard across the instruments.
Please cite this article in press as: J. Vanderwee, et al., The investigation of a relative contrast index model for fingerprint quantification, Forensic Sci. Int. (2010), doi:10.1016/j.forsciint.2010.05.005
G Model
FSI-6081; No. of Pages 6 J. Vanderwee et al. / Forensic Science International xxx (2010) xxx–xxx
5
techniques. The relative contrast index also has a potential application for quality assurance of laboratory reagents. Testing prepared formulations to a standard could also allow the discrimination between discoloured or expired reagents. 5. Conclusions
Fig. 5. Relative contrast index comparison of the depletion groups across different instruments.
This study has indicated that the relative contrast index model is an effective tool for measuring differences in contrast between fingermark samples. Although, the model did not produce absolute or universal values, it still effectively quantified contrast on each instrument. Identifying the model’s applicability to quality assurance in forensic science was also an important outcome. Quantification of fingermark contrast reduces or eliminates ambiguity, as well as providing a documentable, repeatable and objective measurement of contrast enhancement. The relative contrast index model provides a valuable framework and positive outcomes for future forensic science research. Acknowledgements
for. The sampling aperture, operating software and instrument sensitivities perhaps played the key roles in the different results. Technological improvements in the instrument’s design and calibration may compensate for these differences in future. However, current instruments though are somewhat incompatible, which is also evident with different spectral libraries for each different microspectrophotometer, suggesting that the instrumentation cannot produce one standard spectral library [48]. The instruments are capable of comparing samples upon the one instrument but cannot read differences the same way across instruments. The relative contrast index model was found not to produce universal results across the different instruments. The results did not translate into a universal numerical index, such as contrast index models used in photography as measured by densitometers [49]. This was an important development for the model, however, it does not limit its application to the forensic science industry. While the relative contrast index values are not universal across a range of different microspectrophotometers, the results obtained upon the one instrument are still directly comparable. This may provide a quantification of the results obtained from fingermark development research. The comparative results may further be expressed in a percentage of contrast increase or decrease throughout the research results. This empirical quantification may provide a significant advantage when expressing the outcomes of fingermark development and enhancement research. The application of the relative contrast index and other repeatable quality assessment methodologies could potentially improve the quality of enhancement research by standardising the assessment methodologies and providing benchmarks for current
Table 3 Relative contrast index results from the three instruments used in the experiments made from each fingermark depletions. Depletion
Instrument
RCI
Standard deviation
Standard error
n1
1 2 3
0.718 0.489 0.659
0.076 0.078 0.053
0.014 0.014 0.01
n2
1 2 3
0.464 0.296 0.474
0.063 0.035 0.05
0.012 0.006 0.009
n3
1 2 3
0.296 0.191 0.304
0.036 0.022 0.035
0.007 0.004 0.006
The authors would like to thank Professor Chris Lennard and the University of Canberra, Professor James Robertson, Mr Milutin Stoilovic and the Australian Federal Police for their assistance and access to the instruments. References [1] C. Lennard, Conference report fingerprint detection: future prospects, Aust. J. Forensic Sci. 39 (2) (2007) 73–80. [2] J. Almog, Reagents for chemical development of latent fingerprints: vicinal triketones – their reaction with amino acids and with latent fingerprints on paper, J. Forensic Sci. 32 (6) (1987) 1565–1573. [3] J. Almog, A. Hirshfeld, J.T. Klug, Reagents for the chemical development of latent fingerprints: synthesis and properties of some ninhydrin analogues, J. Forensic Sci. 27 (4) (1982) 912–917. [4] J. Almog, V.G. Sears, E. Springer, D.F. Hewlett, S. Walker, S. Wiesner, R. Lidor, E. Bahar, Reagents for the chemical development of latent fingerprints: scope and limitations of benzo[f]ninhydrin in comparison to ninhydrin, J. Forensic Sci. 45 (3) (2000) 538–544. [5] J. Almog, Y. Cohen, M. Azoury, T.R. Hahn, Genipin – a novel fingerprint reagent with colormetric and fluorogenic activity, J. Forensic Sci. 49 (2) (2004) 1367–1371. [6] J. Almog, G. Levinton-Shamuilov, Y. Cohen, M. Azoury, Fingerprint reagents with dual action: color and fluorescence, J. Forensic Sci. 52 (2) (2007) 330–334. [7] J. Almog, A. Klein, I. Davidi, Y. Cohen, M. Azoury, M. Levin-Elad, Dual fingerprint reagents with enhanced sensitivity: 5-methoxy- and 5-methylthioninhydrin, J. Forensic Sci. 53 (2) (2008) 364–368. [8] C. Conn, G. Ramsay, C. Roux, C. Lennard, The effect of metal salt treatment on the photoluminescence of DFO-treated fingerprints, Forensic Sci. Int. 116 (2001) 117–123. [9] S.J. Gardner, D.F. Hewlett, Optimization and initial evaluation of 1,2-indandione as a reagent for fingerprint detection, J. Forensic Sci. 48 (2003) 1288–1292. [10] C.A. Pounds, R. Grigg, T. Mongkolaussavaratana, The use of 1,8-diazafluoren-9one (DFO) for the fluorescent detection of latent fingerprints on paper. a preliminary evaluation, J. Forensic Sci. 35 (1) (1990) 169–175. [11] C. Roux, N. Jones, C. Lennard, M. Stoilovic, Evaluation of 1,2-indanedione and 5,6dimethoxy-1,2-indanedione for the detection of latent fingerprints on porous surfaces, J. Forensic Sci. 45 (4) (2000) 761–769. [12] Y. Sasson, J. Almog, Chemical reagents for the development of latent fingerprints. I: scope and limitations of the reagent 4-dimethylamino-cinnamaldehyde, J. Forensic Sci. 23 (4) (1978) 852–855. [13] L. Schwarz, I. Frerichs, Advanced solvent-free application of ninhydrin for detection of latent fingerprints on thermal paper and other surfaces, J. Forensic Sci. 47 (6) (2002) 1274–1277. [14] L. Schwarz, I. Klenke, Enhancement of ninhydrin- or DFO-treated latent fingerprints on thermal paper, J. Forensic Sci. 52 (3) (2007) 649–655. [15] C. Wallace-Kunkel, C. Lennard, M. Stoilovic, C. Roux, Optimisation and evaluation of 1,2-indanedione for use as a fingermark reagent and its application to real samples, Forensic Sci. Int. 168 (2007) 14–26. [16] A. Becue, C. Champod, P. Margot, Use of gold nanoparticles as molecular intermediates for the detection of fingermarks, Forensic Sci. Int. 168 (2007) 169–176. [17] N. Jones, D. Mansour, M. Stoilovic, C. Lennard, C. Roux, The influence of polymer type, print donor and age on the quality of fingerprints developed on plastic substrates using vacuum metal deposition, Forensic Sci. Int. 124 (2001) 167–177. [18] N. Jones, M. Stoilovic, C. Lennard, C. Roux, Vacuum metal deposition: developing latent fingerprints on polyethylene substrates after the deposition of excess gold, Forensic Sci. Int. 123 (2001) 5–12.
Please cite this article in press as: J. Vanderwee, et al., The investigation of a relative contrast index model for fingerprint quantification, Forensic Sci. Int. (2010), doi:10.1016/j.forsciint.2010.05.005
G Model
FSI-6081; No. of Pages 6 6
J. Vanderwee et al. / Forensic Science International xxx (2010) xxx–xxx
[19] N. Jones, M. Stoilovic, C. Lennard, C. Roux, Vacuum metal deposition: factors affecting normal and reverse development of latent fingerprints on polyethylene substrates, Forensic Sci. Int. 115 (2001) 73–88. [20] B. Schnetz, P. Margot, Technical note: latent fingermarks, colloidal gold and multimetal deposition (MMD) optimisation of the method, Forensic Sci. Int. 118 (2001) 21–28. [21] E. Stauffer, A. Becue, K.V. Singh, K.R. Thampi, C. Champod, P. Margot, Single-metal deposition (SMD) as a latent fingermark enhancement technique: an alternative to multimetal deposition (MMD), Forensic Sci. Int. 168 (2007) e5–e9. [22] M.J. Choi, T. Smoother, A.A. Martin, A.M. McDonagh, P.J. Maynard, C. Lennard, C. Roux, Fluorescent TiO2 powders prepared using a new perylene diimide dye: applications in latent fingermark detection, Forensic Sci. Int. 173 (2007) 154–160. [23] J.D. James, C.A. Pounds, B. Wilshire, Flake metal powders for revealing latent fingerprints, J. Forensic Sci. 36 (5) (1991) 1368–1375. [24] L. Liu, S.K. Gill, Y. Gao, L.J. Hope-Weeks, K.H. Cheng, Exploration of the use of novel SiO2 nanocomposites doped with fluorescent eu3+/sensitizer complex for latent fingerprint detection, Forensic Sci. Int. 176 (2008) 163–172. [25] L.K. Seah, U.S. Dinish, W.F. Phang, Z.X. Chao, V.M. Murukeshan, Fluorescence optimisation and lifetime studies of fingerprints treated with magnetic powders, Forensic Sci. Int. 152 (2005) 249–257. [26] B.J. Theaker, K.E. Hudson, F.J. Rowell, Doped hydrophilic silica nano- and microparticles as novel agents for developing latent fingerprints, Forensic Sci. Int. 174 (2008) 26–34. [27] I. Bialek, J. Brzozowski, A. Lukasik, Comparison of the effectiveness of selected methods of contrasting of fingerprints in blood, Z Zagadnien Nauk Sadowych (Problems of Forensic Science) 59 (2004) 50–65. [28] B. Marchant, C. Tague, Developing fingerprints in blood: a comparison of several chemical techniques, J. Forensic Ident. 57 (2007) 76–93. [29] V.G. Sears, C.P.G. Butcher, T.M. Prizeman, Enhancement of fingerprints in blood – Part 2: protein dyes, J. Forensic Ident. 51 (2001) 28–38. [30] V.G. Sears, T.M. Prizeman, Enhancement of fingerprints in blood – Part 1: the optimisation of amido black, J. Forensic Ident. 50 (2000) 470–480. [31] M. Stoilovic, Improved method for DFO development of latent fingerprints, Forensic Sci. Int. 60 (1993) 141–153. [32] C. Bersellini, L. Garofano, M. Giannetto, F. Lusardi, G. Mori, Development of latent fingerprints on metallic surfaces using electropolymerization processes, J. Forensic Sci. 46 (4) (2001) 871–877. [33] D.T. Burns, J.K. Brown, A. Dinsmore, K.K. Harvey, Base-activated latent fingerprints fumed with a cyanoacrylate monomer. A quantitative study using fourier-transform infra-red spectroscopy, Anal. Chim. Acta 362 (1998) 171–176.
[34] N.J. Crane, E.G. Bartick, R. Schwartz Perlman, S. Huffman, Infrared spectroscopic imaging for noninvasive detection of latent fingerprints, J. Forensic Sci. 52 (1) (2007) 48–53. [35] P. Cuce`, G. Polimeni, A.P. Lazzaro, G. De Fulvio, Small particle reagents technique can help to point out wet latent fingerprints, Forensic Sci. Int. 146S (2004) S7–S8. [36] J. Deans, Recovery of fingerprints from fire scenes and associated evidence, Sci. Justice 46 (2006) 153–168. [37] E. Halahmi, O. Levi, L. Kronik, R.L. Boxman, Development of latent fingerprints using a corona discharge, J. Forensic Sci. 42 (5) (1997) 833–841. [38] O.P. Jasuja, G.D. Singh, G.S. Sodhi, Development of latent fingerprints on compact disc and its effect on subsequent data recovery, Forensic Sci. Int. 156 (2006) 237– 241. [39] Z. Jian, G. Dao-an, A modified cyanoacrylate technique utilizing treated neutral filter paper for developing latent fingerprints, Forensic Sci. Int. 52 (1991) 31–34. [40] J.B. Kempton, W.F. Rowe, Contrast enhancement of cyanoacrylate-developed latent fingerprints using biological stains and commercial fabric dyes, J. Forensic Sci. 37 (1) (1992) 99–105. [41] K. Kent, M. Stoilovic, Development of latent fingerprints using preferential dc sputter deposition, Forensic Sci. Int. 72 (1995) 35–42. [42] Y. Migron, D. Mandler, Development of latent fingerprints on unfired cartridges by palladium deposition: a surface study, J. Forensic Sci. 42 (6) (1997) 986–992. [43] D. Porta, M. Maldarella, M. Grandi, C. Cattaneo, A new method of reproduction of fingerprints from corpses in a bad state of preservation using latex, J. Forensic Sci. 52 (6) (2007) 1–3. [44] M. Tahtouh, P. Despland, R. Shimmon, J.R. Kalman, B.J. Reedy, The application of infrared chemical imaging to the detection and enhancement of latent fingerprints: method optimization and further findings, J. Forensic Sci. 52 (5) (2007) 1089–1096. [45] J.D. Wilson, A.A. Cantu, G. Antonopoulos, M.J. Surrency, Examination of the steps leading up to the physical developer process for developing fingerprints, J. Forensic Sci. 52 (2) (2007) 320–329. [46] C.G. Worley, S.S. Wiltshire, T.C. Miller, G.J. Havrilla, V. Majidi, Detection of visible and latent fingerprints using micro-X-ray fluorescence elemental imaging, J. Forensic Sci. 51 (1) (2006) 57–63. [47] J.D. Humphreys, G. Porter, M. Bell, The quantification of fingerprint quality using a relative contrast index, Forensic Sci. Int. 178 (2008) 46–53. [48] R. Saferstein, Criminalistics:, An Introduction to Forensic Science, Eighth ed., Pearson Prentice Hall, New Jersey, 2004. [49] M. Langford, Basic Photography, Seventh ed, Focal Press, Oxford, 2000.
Please cite this article in press as: J. Vanderwee, et al., The investigation of a relative contrast index model for fingerprint quantification, Forensic Sci. Int. (2010), doi:10.1016/j.forsciint.2010.05.005
